Exploration on Visualization of Uncertainty
Another week of reading about visualization of uncertainty. The more I explored, the more I found it intriguing.
Studies has shown that most participants agreed that display of uncertainty helped with decision making [1,4] and preferred uncertainty to be displayed integratedly rather than separately [1,2]. Depending on which concept is being used, uncertainty can be described as scalars (often used for confidence levels, errors, differences or likelihood), pairs (often used for intervals, ranges, mean or standard deviation) or distributions (often used if sufficient sampling is available) .
Uncertainty can be conveyed in different forms including glyph, modified contour, transparency, fog, texture, box plot, animation, histogram, boundary fuzziness, and color saturation, etc [1,2,3,4]. Among all the forms, color saturation, color lightness and boundary fuzziness were proved to be powerful in representation of uncertainty [1,2,3].
Visualizing uncertainty in ensemble data has been a hot spot for a long time, especially in hurricane forecast. A lot of researchers believe that compared to the traditional cone of uncertainty, ensemble visualization can provide a better overall understanding of the uncertainty and unpredictability associated with forecasted hurricane tracks  and help reduce the chance of misinterpreting the center of the tracks as the hurricane center . But studies has also shown that ensemble visualization could be more difficult to work with .
As for color usage, lighter colors were proved that they were more associated with less certainty. Red and orange were more preferred for uncertain information while green was more easily to be perceived more certain . Yellow, on the other hand, is a tricky one. According to a previous study, the interpretation of it depended on the context. When it was presented with red and orange, it was more preferred for certain information while with green and blue, it was more likely to be considered as a representation of uncertain .
1. Aerts, J. C. J. H., Clarke, K. C., & Keuper, A. D. (2003). Testing Popular Visualization Techniques for Representing Model Uncertainty. Cartography and Geographic Information Science, 30(3), 249–261. doi: 10.1559/152304003100011180
2. Edwards, L. D., & Nelson, E. S. (2001). Visualizing Data Certainty: A Case Study Using Graduated Circle Maps. Cartographic Perspectives, (38), 19–36. doi: 10.14714/cp38.793
3. Pang, A. 2001. Visualizing uncertainty in geo-spatial data. In: Proceedings of the Workshop on the Intersections between Geospatial Information and Information Technology. National Academies Committee of the Computer Science and Telecommunications Board, Washington, D.C.
4. Scholz, R., & Lu, Y. (2014). Uncertainty in Geographic Data on Bivariate Maps: An Examination of Visualization Preference and Decision Making. ISPRS International Journal of Geo-Information, 3(4), 1180–1197. doi: 10.3390/ijgi3041180
5. Cox, J., House, D., & Lindell, M. (2013). Visualizing Uncertainty In Predicted Hurricane Tracks. International Journal for Uncertainty Quantification, 3(2), 143–156. doi: 10.1615/int.j.uncertaintyquantification.2012003966
6. Liu, L., Boone, A. P., Ruginski, I. T., Padilla, L., Hegarty, M., Creem-Regehr, S. H., … House, D. H. (2017). Uncertainty Visualization by Representative Sampling from Prediction Ensembles. IEEE Transactions on Visualization and Computer Graphics, 23(9), 2165–2178. doi: 10.1109/tvcg.2016.2607204
7. Susanne Tak and Alexander Toet. 2014. Color and Uncertainty: It is not always Black and White. Eurographics Conference on Visualization (2014), 55–59. DOI:http://dx.doi.org/10.2312/eurovisshort.20141157